| Object Prevalence |
Article Index for Object |
Website Links For Object |
Information AboutObject Prevalence |
|
In prevalence model, the data the application is using is kept in Memory , rather than being stored in RDBMS or some other common data storage system. A snapshot of data is regularly saved to disk, and in addition to this, all changes are serialised and a log of transactions is also stored on disk. The system is able to provide ACID transaction reliability. The advantage of the system is that simply keeping objects in memory in their normal, natural format is both faster and more programmer-friendly than the multiple conversions that are needed when the objects are stored and retrieved from an RDBMS. The obvious downside is that the application needs a lot of memory, though this is somewhat alleviated by decreasing RAM prices, and the fact that many business databases are small enough to fit in memory anyway. Other downside is that there is no storage standard and each prevalence framework uses its own serialisation format, often undecipherable without knowledge of the implementation details. IMPLEMENTATIONS The pioneer of the method in modern use is Prevayler , a prevalence framework for Java , introduced in 2001 by Klaus Wuestefeld . He does admit he didn't invent the method of journaled transactions as such, but he did coin the object prevalence term as it is used now. Since then, prevalence frameworks have appeared on many languages. Other examples include
A complete list can be found at the Prevayler Wiki. EXTERNAL LINKS |
|
|